Citation

Wadood, Arbab Sufyan and Sajid, Ahthasham and Alam, Muhammad Mansoor and Su’ud, Mazliham Mohd and Mehmood, Arshad and Khan, Inam Ullah (2025) Building Detection from Satellite Imagery Using Morphological Operations and Contour Analysis over Google Maps Roadmap Outlines. International Journal of Advanced Computer Science and Applications, 16 (1). ISSN 2158107X

Text Paper.pdf - Published Version Restricted to Repository staff only Download (3MB)

Abstract

One such research area is building detection, which has a high influence and potential impact in urban planning, disaster management, and construction development. Classifying buildings using satellite images is a difficult task due to building designs, shapes, and complex backgrounds which lead to occlusion between buildings. The current study introduces a new method for constructing recognition and classification globally based on Google Maps contour trace detection and an evolved image processing technique, seeking synergies with a systematic methodology. We first extract the building outlines by taking the image from the ¨Roadmap¨view in Google Maps, converting it to gray scale, thresholding it to create binary boundaries,and finally applying morphological operations to facilitate noise removal and gap filling. These binary outlines are overlaid on colorful satellite imagery, which aids in identifying buildings. Machine learning techniques can also be used to improve aspect ratio analysis and improve overall detection accuracy and performance.

Item Type:	Article
Uncontrolled Keywords:	Building detection; satellite imagery; urban planning; disaster response; image processing; machine learning; morphological operations; contour detection; aspect ratio
Subjects:	Q Science > QA Mathematics > QA299.6-433 Analysis
Divisions:	Faculty of Computing and Informatics (FCI)
Depositing User:	Ms Suzilawati Abu Samah
Date Deposited:	06 Mar 2025 01:24
Last Modified:	06 Mar 2025 01:24
URII:	http://shdl.mmu.edu.my/id/eprint/13579

Downloads

Edit (login required)